Focal segmental glomerulosclerosis (FSGS) and minimal change disease (MCD) are a group of rare diseases that cause serious morbidity and high mortality. There is growing consensus that the presently employed histopathology-based classification of FSGS/MCD is limited because it is not based on an understanding of the molecular basis of these diseases, and because it does not well predict the heterogeneous natural history or response to therapy of those affected. Given these shortcomings, it is not surprising that our therapeutic approach to these diseases is imperfect. We propose that several major barriers must be overcome before more effective interventional studies of primary non-inflammatory glomerular disease can be conducted. Among these barriers is the absence of specific biomarkers of glomerular disease that would allow refined, biologically relevant sub-classification of glomerular disease histopathology useful for defining subject inclusion and exclusion criteria in clinical studies. Such disease sub-classification might overcome the effects of study population heterogeneity that likely have complicated interpretation of past studies of these glomerular diseases. New glomerular disease biomarkers might also predict disease natural history, allow proper selection of and prediction of response to specific therapeutic intervention, allow early detection of disease, or provide indicators of disease activity. Importantly, a robust investigative infrastructure is presently lacking that would facilitate collection, cultivation, and access to human biological material and associated clinical data necessary for biomarker identification, for the identification of clinically relevant study endpoints, and for conducting pilot clinical studies that would advance the care of FSGS/MCD patients. This application proposes a collaborative prospective observational cohort study of patients who present with histopathology characteristic of FSGS/MCD. After collecting 250 patients with this kidney biopsy characteristic, and their associated clinical data, kidney tissue, blood, and urine, the initial goals of the FSGS/MCD Cohort study are: (a) to develop and use a combination of molecular phenotypes, quantifiable histological parameters, and discrete clinical features to predict clinical outcomes; and (b) to classify patients according to their molecular phenotype into discrete subgroups. These studies hold out the promise to the clinician of being able to predict the natural history of FSGS/MCD.